diff --git a/swh/model/cli.py b/swh/model/cli.py
index 5996d19..6df219e 100644
--- a/swh/model/cli.py
+++ b/swh/model/cli.py
@@ -1,95 +1,102 @@
# Copyright (C) 2018 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU General Public License version 3, or any later version
# See top-level LICENSE file for more information
import click
+import locale
import os
import sys
from swh.model import identifiers as pids
from swh.model.exceptions import ValidationError
from swh.model.from_disk import Content, Directory
class PidParamType(click.ParamType):
name = 'persistent identifier'
def convert(self, value, param, ctx):
try:
pids.parse_persistent_identifier(value)
return value # return as string, as we need just that
except ValidationError as e:
self.fail('%s is not a valid PID. %s.' % (value, e), param, ctx)
def pid_of_file(path):
object = Content.from_file(path=path).get_data()
return pids.persistent_identifier(pids.CONTENT, object)
def pid_of_dir(path):
object = Directory.from_disk(path=path).get_data()
return pids.persistent_identifier(pids.DIRECTORY, object)
@click.command()
@click.option('--type', '-t', default='auto',
type=click.Choice(['auto', 'content', 'directory']),
help='type of object to identify (default: auto)')
@click.option('--verify', '-v', metavar='PID', type=PidParamType(),
help='reference identifier to be compared with computed one')
+@click.option('--filename/--no-filename', 'show_filename', default=True,
+ help='show/hide file name (default: show)')
@click.argument('object',
type=click.Path(exists=True, readable=True,
allow_dash=True, path_type=bytes))
-def identify(type, verify, object):
+def identify(type, verify, show_filename, object):
"""Compute the Software Heritage persistent identifier (PID) for a given
source code object.
For more details about Software Heritage PIDs see:
\b
https://docs.softwareheritage.org/devel/swh-model/persistent-identifiers.html
\b
Examples:
\b
$ swh-identify /usr/src/linux/kernel/
swh:1:dir:f9f858a48d663b3809c9e2f336412717496202ab
\b
$ swh-identify /usr/src/linux/kernel/sched/deadline.c
swh:1:cnt:57b939c81bce5d06fa587df8915f05affbe22b82
"""
if type == 'auto':
if os.path.isfile(object):
type = 'content'
elif os.path.isdir(object):
type = 'directory'
else: # shouldn't happen, due to path validation
raise click.BadParameter('%s is neither a file nor a directory' %
object)
pid = None
if type == 'content':
pid = pid_of_file(object)
elif type == 'directory':
pid = pid_of_dir(object)
else: # shouldn't happen, due to option validation
raise click.BadParameter('invalid object type: ' + type)
if verify:
if verify == pid:
click.echo('PID match: %s' % pid)
sys.exit(0)
else:
click.echo('PID mismatch: %s != %s' % (verify, pid))
sys.exit(1)
else:
- click.echo(pid)
+ msg = pid
+ if show_filename:
+ encoding = locale.getpreferredencoding(do_setlocale=False)
+ msg = '%s\t%s' % (pid, object.decode(encoding))
+ click.echo(msg)
if __name__ == '__main__':
identify()
diff --git a/swh/model/tests/test_cli.py b/swh/model/tests/test_cli.py
index 054cc0c..d8e68b9 100644
--- a/swh/model/tests/test_cli.py
+++ b/swh/model/tests/test_cli.py
@@ -1,73 +1,103 @@
# Copyright (C) 2018 The Software Heritage developers
# See the AUTHORS file at the top-level directory of this distribution
# License: GNU General Public License version 3, or any later version
# See top-level LICENSE file for more information
import os
import tempfile
import unittest
from click.testing import CliRunner
from nose.plugins.attrib import attr
from swh.model import cli
from swh.model.tests.test_from_disk import DataMixin
from swh.model.hashutil import hash_to_hex
@attr('fs')
class TestIdentify(DataMixin, unittest.TestCase):
def setUp(self):
super().setUp()
self.runner = CliRunner()
def test_content_id(self):
+ """identify file content"""
self.make_contents(self.tmpdir_name)
for filename, content in self.contents.items():
path = os.path.join(self.tmpdir_name, filename)
result = self.runner.invoke(cli.identify,
['--type', 'content', path])
self.assertEqual(result.exit_code, 0)
- self.assertEqual(result.output.rstrip(),
+ self.assertEqual(result.output.split()[0],
'swh:1:cnt:' + hash_to_hex(content['sha1_git']))
def test_directory_id(self):
+ """identify an entire directory"""
self.make_from_tarball(self.tmpdir_name)
path = os.path.join(self.tmpdir_name, b'sample-folder')
result = self.runner.invoke(cli.identify,
['--type', 'directory', path])
self.assertEqual(result.exit_code, 0)
- self.assertEqual(result.output.rstrip(),
+ self.assertEqual(result.output.split()[0],
'swh:1:dir:e8b0f1466af8608c8a3fb9879db172b887e80759')
+ def test_show_filename(self):
+ """filename is shown by default"""
+ self.make_contents(self.tmpdir_name)
+ for filename, content in self.contents.items():
+ path = os.path.join(self.tmpdir_name, filename)
+ result = self.runner.invoke(cli.identify,
+ ['--type', 'content', path])
+
+ self.assertEqual(result.exit_code, 0)
+ self.assertEqual(result.output.rstrip(),
+ 'swh:1:cnt:%s\t%s' %
+ (hash_to_hex(content['sha1_git']), path.decode()))
+
+ def test_hide_filename(self):
+ """filename is hidden upon request"""
+ self.make_contents(self.tmpdir_name)
+ for filename, content in self.contents.items():
+ path = os.path.join(self.tmpdir_name, filename)
+ result = self.runner.invoke(cli.identify,
+ ['--type', 'content', '--no-filename',
+ path])
+
+ self.assertEqual(result.exit_code, 0)
+ self.assertEqual(result.output.rstrip(),
+ 'swh:1:cnt:' + hash_to_hex(content['sha1_git']))
+
def test_auto_id(self):
+ """automatic object type: file or directory, depending on argument"""
with tempfile.NamedTemporaryFile(prefix='swh.model.cli') as f:
result = self.runner.invoke(cli.identify, [f.name])
self.assertEqual(result.exit_code, 0)
self.assertRegex(result.output, r'^swh:\d+:cnt:')
with tempfile.TemporaryDirectory(prefix='swh.model.cli') as dirname:
result = self.runner.invoke(cli.identify, [dirname])
self.assertEqual(result.exit_code, 0)
self.assertRegex(result.output, r'^swh:\d+:dir:')
def test_verify_content(self):
+ """identifier verification"""
self.make_contents(self.tmpdir_name)
for filename, content in self.contents.items():
expected_id = 'swh:1:cnt:' + hash_to_hex(content['sha1_git'])
# match
path = os.path.join(self.tmpdir_name, filename)
result = self.runner.invoke(cli.identify,
['--verify', expected_id, path])
self.assertEqual(result.exit_code, 0)
# mismatch
with open(path, 'a') as f:
f.write('trailing garbage to make verification fail')
result = self.runner.invoke(cli.identify,
['--verify', expected_id, path])
self.assertEqual(result.exit_code, 1)